Research Analyzer
← Back ICRA 2026

Friction-Aware Actuator Modeling for Accurate Torque Estimation Using External Sensors

Jiman Park, Hyunyong Lee, Hansol Kang, SeongWon Nam, Yeongwoo Son, Bumsu Yi, Jaeyoung Oh, Hyouk Ryeol Choi

PDF

AI summary

Key figure (auto-extracted from paper)
A friction-compensation method using external torque sensors accurately isolates pure actuator torque, effectively bridging the sim-to-real gap.
Actuator modeling torque estimation friction compensation sim-to-real external sensors robotics

Problem

External torque sensors are distorted by disturbances from external load systems, making it difficult to measure an actuator's pure output torque and causing sim-to-real discrepancies.

Approach

The method identifies baseline friction under no-load conditions using a Stribeck model and compensates loaded sensor measurements to isolate the true actuator torque constant.

Key results

  • Identifies friction current as a function of angular velocity using a Stribeck model
  • Compensates load-induced friction to stabilize the estimated torque constant across varying currents
  • Achieves torque estimation with under 3.0% RMSE across multiple actuators
  • Maintains high accuracy even in actuators with significant sealing friction

Why it matters

Provides a practical, data-efficient solution for accurate actuator modeling that reduces sim-to-real gaps across diverse robotic systems.

Abstract

Modern robotic controllers are typically designed in simulation and subsequently deployed on real robots. How- ever, discrepancies between simulated and actual actuator torque often lead to sim-to-real (sim2real) problems. Vari- ous actuator approaches have been proposed to address this problem, but when external torque sensors are used, it is difficult to measure the intrinsic actuator output torque due to disturbances from external load systems. This paper proposes an actuator modeling method that minimizes the influence of external systems. The friction torque of the actuator is first identified under no-load conditions, and the measured torque under loaded conditions is compensated accordingly to estimate the pure output torque. Experimental results across various actuators and load conditions demonstrate that the proposed model closely matches the measured torque, even in actuators with large friction. The proposed approach overcomes the modeling limitation using external sensors and provides an effective solution for reducing sim2real problems in diverse actuator systems.

Index terms

Dynamics Software-Hardware Integration for Robot Systems Calibration and Identification

Related papers